Chapter 6 - Springer Link

Chapter 6 Urban Diversity and Intersectoral Diffusion Some Insightsfrom the Study ofTechnical Creativity PIERRE DESROCHERS Montreal Economic Institute

1.

INTRODUCTION

Probably the least controversial aspect of technological change is that people solve problems by combining heterogeneous facts, ideas, faculties, and skills (Koestler 1969; Usher 1966). The spread of technical information from one industrial milieu to another has therefore long been recognized as a significant element in the development of technology (Langrish et at. 1972: 43). This process is everywhere to be found, for most materials, tools and products always end up with numerous uses different from those intended originally (Basalla 1988; Rosenberg 1994, 1996; Smith 1982). For example, practical metallurgy began with the making of necklace beads and ornaments in hammered native copper long before knives and weapons were made. Ceramics began with the fire-hardening of fertility figurines molded in clay, while rotary motion was first used in the drilling and shaping of necklace beads (Smith 1982). In the last century, the concept of a production chain was adapted in flourmills, slaughterhouses, and machine tool, canning, railroad and car assembly factories (Hounshell 1991; Klemm 1959; Mokyr 1990). Meanwhile, steel was progressively adopted in railroad and naval construction, high-rise buildings, and steam engines, among other areas. Laser, which was for a long time a solution in search of a problem, is now used in, among other things, printers, telecommunication equipment, navigational instruments, textile machinery, optic surgery, precision measurement, and weapon systems (Rosenberg 1996). Despite the obvious fact that innovations continually cross so-called "industry boundaries, regional specialization has long been held by most M. P. Feldman et al. (eds.), Institutions and Systems in the Geography of Innovation © Springer Science+Business Media New York 2002

100

Institutions and Systems in the Geography ofInnovation

urban economists and economic geographers as the optimal setting to promote innovation. One of the few authors to strongly dissent from that view is Jacobs (1969), who argues, among other things, that local diversity increases the probability of combining different resources. In a pioneering paper, Glaeser et al. (1992) attempted to assess empirically the respective impact of local specialization and diversity on economic growth gave support to Jacobs's hypothesis and started a controversy on the type of localized knowledge spillovers (intra- or inter-industrial) that are more conducive to innovation (Feldman 2000). If this debate has generated much measurement, however, a clear theoretical understanding of the processes by which knowledge "spills over" from a particular application domain to other industrial sectors and the role that geographical proximity might play in these processes has still not been provided. As such, many recent empirical studies suffer from problems similar to those that plagued earlier work on "interindustry technology flows" (DeBresson 1990, 1996) and the measurement of local diversity (Jackson 1984). It is therefore the purpose of this chapter to make, in a preliminary way, these processes more explicit.

2.

SPECIALIZATION VERSUS DIVERSITY IN THE LOCALIZED TRANSMISSION OF INTERINDUSTRY KNOWLEDGE SPILLOVERS: SOUND AND FURY OVER WHAT?

In recent years, there has been a heated debate as to which is better conducive to technological change and economic growth: local diversity or specialization of economic activity (Glasmeier 2000). While some authors argue that local economic diversity is more conducive to "interindustry knowledge spillovers" and subsequent economic growth (Feldman and Audretsch 1999; Glaeser et al. 1992; Harrison, Kelley, and Gant 1996a, 1996b), others claim more modestly that diversified local economies generate more employment growth than the more homogeneous ones (Coffey and Shearmur 1998). On the other hand, some authors argue that while localized diversity might be important in certain cases, local specialization is generally more conducive to innovation and growth (Bostic, Gans, and Stem 1997; Henderson 1997; Henderson, Kuncoro, and Turner 1995). While these studies raise important questions and provide innovative approaches to further our understanding of regional growth, they have been the subjects of a number of criticisms. For example, Glaeser et al. (1992) use wages and employment growth as dependent variables, assuming that they are measurable effects of innovation and new knowledge when actually

Urban Diversity and Intersectoral Diffusion

101

these results simply cannot be interpreted in terms of knowledge and education (Quigley 1998: 136). Other studies use indicators, such as innovation citations (Feldman and Audretsch 1999) and particular technology adoption (Harrison, Kelley, and Gant 1996a, 1996b), that seem more adequate but that nonetheless have some problems of their own, as their authors point out in some detail. Perhaps, however, the main problem confronting these research designs is that they tend to make dynamic inferences from cross-sectional data (Malmberg 1997) and that they do not adequately address the processes by which materials, production processes, and products are adapted in many different activities. Most of them deal casually with the topic by referring to "Jacobs' externalities," which is the argument that the spatial concentration of diverse individuals permits a great deal of personal interaction, which in turn generates new ideas, products, and processes. Consider, for example, the short treatment given to these processes in the most important studies on the topic: Jacobs' idea is that the crucial externality in cities is cross-fertilization of ideas across different lines of work. New York grain and cotton merchants saw the need for national and international financial transactions, and so the financial services industry was born. A San Francisco food processor invented equipment leasing when he had trouble finding financing for his own capital: the industry was not invented by bankers. In a more systematic account, Rosenberg (1976) discusses the spread of machine tools across industries and describes how an idea is transmitted from one industry to another. Scherer (1982) presents systematic evidence indicating that around 70 percent of inventions in a given industry are used outside that industry. Much evidence thus suggests that knowledge spills over across industries. Because cities bring together people from different walks of life, they foster transmission of ideas. (Glaeser, et al. 1992: 1131 - 1132) Jacobs' effects are supposed to derive from the diversity of the local urban environment that surrounds an industry. Diversity enhances knowledge accumulation as producers in an industry can draw upon a greater diversity of ideas from other industries, through interacting socially and commercially. (Henderson 1997: 464) Jacobs (1969) argues that the most important sources of knowledge spillovers are external to the industry in which the firm operates and that cities are the source of innovation because the diversity of these knowledge sources is greatest in cities. Thus Jacobs develops a theory that emphasizes that the variety of industries within a geographic region

102


promotes knowledge externalities and ultimately innovative activity and economic growth. (Feldman and Audretsch 1999: 412) Jacobs's work, however, does not provide a systematic description of these transfers, although she hints at a number of processes through the use of illustrative anecdotes. Furthermore, she makes no clear distinction between knowledge spillovers on the one hand and traditional urbanization externalities and entrepreneurship on the other. Perhaps Jacobs is right in doing so, for in reality the demarcation between these factors is probably never obvious. Yet if one wants to interpret results derived from various proxies in terms of knowledge spillovers, one should at least provide some detail on how previously unrelated pieces of knowledge are combined to create technological innovation. It will now be argued that the best way to do this is by building on some universal features of human creativity.

3.

PROBLEM-SOLVING, COMBINATIONS, AND TECHNOLOGICAL CHANGE

3.1

The Combination Process

All innovations are essentially new combinations of previously unrelated things. As Fores (1979: 853) points out, the main thrust of an engineer or a technician is "to gather knowledge from diverse places in order to help solve technical problems." As the economist Clarence E. Ayres (1943: 113) puts it: "The history of every material is the same. It is one of novel combinations of existing devices and materials in such a fashion as to constitute a new device or a new material or both." Gutenberg's invention of the printing press affords a well-known illustration. At the dawn of the fifteenth century, printing was no longer a novelty in Europe. Printing from wooden blocks on vellum, silk, and cloth apparently started in the twelfth century, and printing on paper was widely practiced in the second half of the fourteenth. Oddly enough, though, the starting point of Gutenberg's invention was playing-cards on which a few words has been printed by way of rubbing wood-blocks on a sheet of paper. As he wrote in his correspondence to a clergyman: Well, what has been done for a few words, for a few lines, I must succeed in doing for large pages of writing, for large leaves covered entirely on both sides, for whole books, for the first of all books, the Bible. How? It is useless to think of engraving on pieces of wood the whole thirteen hundred pages. What am I to do? I do not know: but I


103

know what I want to do: I wish to manifold the Bible; I wish to have the copy ready for the pilgrimage to Aix-la-Chapelle. (Koestler 1969: 122). Gutenberg then searched for a device more resistant than wood block, which lead him to notice the seals used to authenticate documents, but rubbing them on paper did not give a clear print. He found the solution one day, while attending a wine harvest near his city. I took part in the wine harvest. I watched the wine flowing, and going back from the effect to the cause, I studied the power of this press that nothing can resist. God has revealed to me the secret that I demanded of Him. One must strike, cast, make a form like the seal of your community; a mold such as that used for casting your pewter cups; letters in relief like those on your coins, and the punch for producing them like the foot when it multiplies its print. There is the Bible (Koestler 1969: 123-24). The combination of previously unrelated pieces of knowledge and artifacts has important implications that have often been neglected by authors working on interindustry knowledge spillovers and local diversity through the use of conventional classification systems. Once one recognizes the importance of combining diverse resources to produce any kind of innovation, it quickly becomes obvious that the whole conceptual framework of "industrial classification" typically becomes more misleading than useful to understand innovative processes (Jacobs 1969; Rosenberg 1976; Weber 1992). Furthermore, industrial classification schemes will typically hide firms involved in the production of related products or services. For example, Porter (2000: 255) points out that Massachusetts firms involved in the production of medical devices were "buried within several larger and overlapping industry categories, such as electronic equipment and plastic products."

3.2

Natural Versus Artifactual Evolutionary Processes

As many authors have pointed out, the combination of previously unrelated things is the main difference between "natural" and "artifactual" evolution (Basalla 1988; Mokyr 1990; Sahal 1981; Weber 1992). In short, with the exception of the microbe and gene levels, different biological entities usually do not interbreed, while artifactual types are relentlessly combined to produce new entities. The anthropologist Alfred L. Kroeber (1948) illustrated this critical difference between living and human-made things by sketching a "family tree" of organic species and another of cultural artifacts (Figure 1).


104

,~~ . .::~~:~~I~' f'1f~'/" 2 I; l

'~~ IVPV)i I

1,1

:?i~~vf2~ ~~~ '~I! ~~-·'.~l'\fib~< ....

. '.

-

~111"1'~ .1 ,~.y .....~~_.

~'i~r~

• Ji

~

.

:'1

J L Figure 1. Family Trees of Organic Life and Cultural Artifacts Source: Kroeber 1948: 260.

In Kroeber's illustration, the species that form the different branches in the tree of life do not readily mix, but they split to form new species and remain totally isolated from one another once the "speciation" process has been completed. By contrast, the branches of the artifactual tree fuse together to produce new types, which merge again with other branches. It could thus be said that a creative human being once had the idea to "mate" a tree and duck to produce a wooden duck decoy. Another way to look at this is to say that the internal combustion engine branch was joined with that of the bicycle and horse-drawn carriage to create the automobile branch, which in turn merged with the dray wagon to produce the motor truck. In principle, any invention can mate with any other invention, although the issue of commercial success is another matter. And, of course, a researcher can take genes out of a fish and put them in a strawberry. On the other hand, the Standard Industrial Classification (SIC) system and the North American Industry Classification System (NAICS), with their definition of industry as a collection of firms producing a homogeneous product, are embedded in a rather straightforward "speciation" framework of evolutionary change. The problem is that one cannot reconcile the SIC and the NAICS with the fact that virtually all functional processes and materials continually cut across "industrial branches." The concept of "inter-industry" technology flows then becomes superfluous or is, at any rate, incompatible with a speciation view of technological change. I It is therefore not surprising that according to DeBresson (1990: 833), one quickly wonders what is measured and what assumptions and hypotheses are used in the literature on interindustrial knowledge spillovers. Similar problems are encountered in studies on the measurement of local economic diversity. Jackson (1984: 103) thus points out that the issue is typically "swamped by the measurement


105

and estimation techniques employed" and that in the end "current diversity measures are deemed inadequate for regional policy makers." Another problem with SIC and NAICS data is that they hide the multi-product nature of virtually all firms of any significance. The possibility of combinations within firms possessing very different resources and capabilities are therefore ignored, even though such processes occur on a routine basis. For example, some employees of Canon's electronics and optics division have combined their knowledge to create significant innovations in cameras and photocopying machines (Galunic and Rodan 1997), while some of Sharp's employees developed the first commercially viable liquid crystal display for pocket calculators from the fusion of electronic, crystal, and optic technologies (Kodama 1992). Raymond Kurzweil, a pioneer in the field of electronic music synthesizers, describes how people with very different skills work together in his plant: I find that the projects I get involved with are increasingly interdisciplinary in their approach. Inventions like the reading machine are not the kind of product created by someone who goes down into his basement and emerges two years later with some brilliant breakthrough. The projects I work on are a disciplined effort involving teams of people with different backgrounds. One of the key challenges I have in trying to lead these projects is to provide for good communications among people with very different backgrounds. In speech recognition, for example, some of the technologists involved include linguists, signal-processing experts, VLSI [very large-scale integration] designers, psycho-acoustic experts, speech scientists, computer scientists, human-factors designers, experts in artificial intelligence and pattern recognition, and so on. (Brown 1988: 243 - 244) The economic diversity within a given geographical area is therefore typically hidden when medium and large firms with very diversified human resources are assigned a single code. In fact, any notion of diversity that does not deal specifically with the accumulated know-how of human beings ultimately rests on subjective criterions. The very structure of the SIC is a case in point. It has thus long been criticized for, among other things, using both "product" and "production process" criteria to delineate various categories (Economic Classification Policy Committee 1993b) and for ignoring as distinct categories important industries such as plastics and electronics (Economic Classification Policy Committee 1993a). In the end, as Griliches (1990: 1666) points out, it might be that the SIC (and by extension NAICS) is "nothing more than a mirage." A more interesting framework for the issue at hand is the patent classification system that is based on broad technological categories and

106


therefore reflects more adequately the importance of the combination process for "artifactual evolution." Patent data, however, have other important limitations, both in the way they are structured and in the limited coverage of technology that they provide. 2 Their main problem for the combination of resources is that they often do not tell us the "industry of origin" of an invention. Furthennore, inventors and patent office employees can only guess some of the potential uses of a new device or material and the actual or most important use of an invention is often very remote from its initial purpose (Basalla 1988; Jewkes, Sawers, and Stillennan 1969; Smith 1982). Schmookler (1966: 23) was one of the first researchers to struggle with this issue: [A major] deficiency arose from the fact that I could not assign many [patented] inventions to a single industry. In part this resulted from my own ignorance, but often it reflected the interindustry character of technology. Thus, a given improvement in the diesel engine may be used in generating electricity or driving a locomotive, a given bearing may be used in shoemaking machine or a lawn mower, and a given knife may be used in harvesting or in kitchens. In consequence, the patent statistics used below generally do not include power plant inventions, electric motors, bearings, or other instruments or materials whose industry of origin was either multiple or simply not evident. Unfortunately, this means that the railroad data do not include inventions in the field of the steam or diesel engines, and that neither the fann nor the construction data include inventions on tractors. Many analysts have long debated these problems and suggested several solutions, but most of these have proven problematic as well (DeBresson 1996; Desrochers 1998; Griliches 1990). In the end, as Griliches (1990: 1667) points out, "most of the basic questions of classification still remain to be answered." It will therefore be argued that a better approach to understand how local diversity can facilitate the combination of previously unrelated things is to drop the use of classification systems altogether and to focus instead on human creativity.

4.

HUMAN CREATIVITY AND THE PROCESSES OF RESOURCE COMBINATION

In combining resources, an individual uses his previous know-how and his capacity of observation and learning. He therefore has only two ways of combining different resources: (I) he can incorporate a new type of process/product to a previously unrelated process/product; (2) he can find a


107

new use for a process/product. Gutenberg thus already knew how to work metal, a skill that he learned as the child of the Archbishop of Mainz's goldsmith, and it was easy for him to make the transition from wooden to lead moveable. types. It was, however, his observation and subsequent learning about wine presses that provided the final breakthrough in the creation of the first functional printing press. The invention of the printing press was mostly the result of the efforts of one man. Much combination of resources, however, involves the creation of a team of people with different skills or the movement of people among different establishments. Yet even though people often have to work in teams to solve problems, all human minds function separately. As the psychologist Robert Weber (1992: 56) puts it: "Almost all important inventions are the work of multiple minds. But once we extract principles behind their development, it is possible to incorporate those principles into the individual mind, thereby giving us a leg up on the inventive process." One should therefore study individuals rather than firms and regions to grasp the essence of resource combinations. Leaving aside "in-house" resource combinations, one can observe at least four different transmission mechanisms: (1) a firm's employees add to, or switch, their product lines; (2) individuals move from one type of production to another and incorporate their previous know-how in a new activity; (3) individuals observe a product/process in another setting and incorporate it in their production; (4) individuals with different skills working for different firms collaborate with one another to create a new process/product. We shall now look briefly at these processes.

4.1

A Firm's Employees Add to or Switch Their Product Lines

As Weber (1992: 104) has argued, finding new purposes for eXlstmg products and know-how is the "freest lunch that technology can offer," although it is probably often the case that substantial amounts of time and resources might still be needed to achieve a successful result. The point, however, is that this process is found throughout the history of technology. Rosenberg (1976: 53) thus points out a dominant pattern among nineteenthcentury American and British tool manufacturers: "A firm develops a new technique or process in response to a problem in a particular industry -the manufacture of firearms, for example. It later becomes apparent that this technique is applicable to typewriters. The technique is 'transferred' to the typewriter industry by the firm which developed the technique ...undertaking the production of typewriters." This process of adding to production lines is found everywhere. Thus, during the first half of the nineteenth century New

108


York's shipbuilding manufacturers branched out into making carriages, steam engines, and locomotives (Lichtenberg 1960), while in the I890s, numerous buggy, railroad, toy, agricultural equipment, firearms and sewing machine manufacturers turned to the production of bicycles (Hounshell 1991). Crevoisier (1993) points out that in the last few decades, numerous Swiss clock and watch manufacturers have expanded their product lines to such items as surgical tools, pacemakers, pens and insulin pumps, among other things. In some circumstances, however, some skilled employees will have to change jobs to find a new use for their know-how, as will now be argued. 4.1.1

Individuals Move from One type of Production to Another and Incorporate Their Previous Know-How in a New Activity

The movement of skilled personnel between different types of production is conspicuous throughout history. The importance of European workers who had been trained in clock making and then transferred these skills to numerous other activities in the late Middle Ages and early Renaissance is well known. The same phenomenon, but in different industries, is well documented in the early British industrial expansion, most notably through the famous master-apprentices "Bramah" dynasty (Mokyr 1990; Smiles 1967; Thomson 1991). Hounshell (1991) and Hoke (1990) similarly document how mechanics who had learned to create highly productive factories passed those ideas on to others, who expanded and spread them to everything from the fabrication of axes to that of locks, from mechanical reapers to typewriters and sewing machines. The same phenomenon is present every day, as many detailed studies on technological change can attest. Consider, for example, the following illustration given by Langrish et al. (1972: 44): An example of technological development of a new person-joining [a] firm is to be found in English Electric's development of fuses for the protection of semiconductor devices. E. Jacks, then Chief Engineer in the Fusegear Division, had identified the area of printed circuit technology as an area that might provide an answer to manufacture of the fuse elements. Progress was, however, held up because no one in the development team possessed enough skill in the use of photofabrication techniques. This need was overcome when, "by sheer luck," Jacks was interviewing an electrical engineer for a job and the applicant happened to mention casually that he had some skill in industrial photography that he had developed as a hobby. Photofabrication techniques were applied with great success and resulted in a completely novel process in the manufacture of fuse elements.


109

The changes brought about under such circumstances might often look radical to people unfamiliar with the techniques adapted to a new situation, but it is probably fair to say that people who were previously knowledgeable about the transferred techniques are more likely to judge it incremental. Again, applying a particular know-how or material to a new situation might sometimes require considerable developmental efforts.

4.1.2

Individuals Observe a Product/process in Another Setting and Incorporate It In Their Production

Gutenberg's invention of the first functional printing press is a wellknown example of an individual observing a process in another setting and incorporating it in his own invention. Almost equally famous is the case of the car industry. It is thus generally acknowledged that the success of the Ford Motor Company owed much to previous developments in other industries, such as the production of interchangeable parts, the idea of continuous flow, and the rise of an efficiency movement (Hounshell 1991; Klemm 1959). One industry that provided Ford's technical people a model of efficient material handling was the meat packing industry. According to the historian of technology David Hounshell (1991: 241), William Klann, head of the engine department at Ford, recalled touring Swift's Chicago slaughterhouse and suggesting to superintendent P. E. Martin: "If they can kill pigs and cows that way, we can build cars that way and build motors that way." Klann also stressed that the Ford flow production drew on the mechanical conveying system of both the flour milling and brewing industries: "We combined our ideas on the Huetteman & Cramer grain [conveying] machine[ry] experience, and the brewing experience and the Chicago stockyard. They all gave use ideas for our own conveyors" (Hounshell 1991: 241). According to Hounshell, the process technology employed in food canning might also have influenced some of Ford's employees.

4.1.3

Individuals Possessing Different Skills and Working for Different Firms Collaborate to Create a New ProcesslProduct

Exchange of technical information and informal collaboration between the employees of competing firms is widespread (Von HippeJ 1988). It therefore seems likely that collaboration between individuals involved in totally different lines of work would be even easier to accomplish. There is indeed some evidence to that effect. Thus in the 1980s, aerospace manufacturers began using carbon composite material instead of aluminum to make tail sections, wings, noses, and fuselages. Used first in tennis rackets

110


and skis, composite material is just as strong but typically weighs only about half as much as aluminum. But it is also far more expensive and much more difficult to handle because if the composite material is not kept properly refrigerated before being cut to the proper shape, the material will be wasted. Properly refrigerating a huge aircraft while in production, however, was no simple task. When production managers at Northrop began wrestling with this practical dilemma, one of them decided to call on the refrigeration specialists at Sara Lee. Not long after that, knowledge and expertise gathered through decades of refrigerating large facilities became part of modern aircraft production technology (Rothschild 1990). Similar examples are found between individuals working for electronic and fiber-optic firms (Kodama 1992), as well as for electronic and biomedical firms (Miller and Cote 1987), among many others (Desrochers 1999). The importance of combining different resource to create technical innovation is nothing short of overwhelming. It will now be argued that geographical proximity between people possessing very different knowledge bases can facilitate these processes.

5.

GEOGRAPHICAL PROXIMITY AND THE COMBINATION OF RESOURCES

For more than a century, many geographers and economists have developed theories relating to the spatial agglomeration of economic activity in response to three empirical observations: (1) a large portion of world output is produced in a limited number of highly concentrated core regions; (2) firms in similar or related industries tend to co-locate in particular places; (3) both of these patterns seem to be sustainable over time (Malmberg 1997). In the last two decades, however, economic geography as an area of inquiry has moved from essentially being concerned with the two sides of location theory - the cost-oriented analysis of optimum plant location and the demand-oriented analysis of sales areas -to a more qualitative analysis of networks of firms and the regional systems they generate, mostly through locally embedded processes of learning (Malmberg 1996, 1997). Many writers that the geographical concentration of related firms balance cooperative and competitive forms of economic activity, leading to greater innovation and flexibility, while also facilitating new business formation, the development of trust relationships, and easier access to start-up capital, have thus emphasized it. Numerous authors have also pointed out the importance of frequent, face-to-face communication for the transmission of tacit knowledge. In short, geographical deconcentration cannot be accomplished easily for activities dealing with creative work, mainly because long-distance


111

communication is still inadequate for the continuous and detailed engineering or technological adjustments that are needed in the course of innovative work (Von Hippe11988; Desrochers 2001). If a strong case can be made that people sharing a similar knowledge base still need to communicate frequently face-to-face, it seems obvious that such a need will be even greater for people possessing different skills and backgrounds. Because most of the time such individuals do not even share key concepts, there is typically a need to develop a common language to coordinate search and development procedures. Thus Kurzweil explains the difficulty of getting people with very different backgrounds -such as linguists, human-factors designers, computer-scientists, and experts in artificial intelligence -to work together: Each one of these fields has very different methodologies and different terminologies. Very often a term in one field means something else entirely in another field. Sometimes we even create our own terminology for a particular project. So, enabling a team like that to communicate and solve a problem is a significant challenge. If you look at the entire company, you bring in even more disciplines: manufacturing, materialresources planning, purchasing, marketing, finance, and so on. Each of these areas has also developed sophisticated methodologies of their own that are as complex as those in engineering. My challenge is to provide a climate in which people with different expertise can work together toward a common goal and communicate clearly with one another. (Brown 1988: 243 - 44) Feldman (1994: 21) makes similar comments by pointing out that individuals with different expertise have different cognitive schemata. Interpreting and synthesizing information in this context involves constant questioning and interpretation through a process of trial, feedback, and evaluation that is facilitated by face-to-face communication. In a recent work, Desrochers '(1999) provides qualitative evidence on numerous instances of resource combinations between individuals with different backgrounds, illustrating that the importance of this phenomenon is just as great in "low-tech" industries as in more complex endeavors. If there is thus much evidence pointing toward the importance of geographical proximity for communication between people sharing common cognitive ground, it seems obvious that its importance is even greater for people possessing diverse backgrounds.

112

6.


CONCLUSION

One of the least controversial aspects of technological change is that innovation proceeds by the combination of previously unrelated things. Yet in a recent review of the literature on geographically localized knowledge spillovers, Feldman (2000: 389) writes that "we do not know how social interaction is initiated, how it evolves into a working relationship and how economically useful knowledge is created." A logical point of departure for researchers interested in those issues would be to follow the argument of Aitken (1985: 15), who says in his history of the radio that "the points of confluence of information flows define the social locations where there is a high probability of new combinations being made. It is a matter of probabilities: the probabilities of new combinations being formed is higher at the points of confluence of information flows than it is elsewhere." With a few exceptions, the most prominent being Jacobs (1969, 1984), this point has not been thoroughly appreciated so far. There are, however, many reasons to believe that local diversity might be more conducive to innovation and growth. A proper understanding of the processes by which creative individuals combine existing resources in different ways will, however, require that old frames of mind and familiar research designs be discarded.

REFERENCES Aitken, H. J. (1985). The Continuous Wave: Technology and American Radio, 1900 - 1932. Princeton, NJ: Princeton University Press. Ayres, C. E. (1943). The Theory 0/ Economic Progress. Chapel Hill, NC: University of North Carolina Press. Basalla, G. (1988). The Evolution o/Technology. Cambridge: Cambridge University Press. Bostic, R. W., J. S. Gans, and S. Stern. (1997). "Urban Productivity and Factor Growth in the Late Nineteenth Century." Journal o/Urban Economics 41: 38 - 55. Brown, K. A. (1988). Inventors at Work. Interviews with Sixteen Notable American Inventors. Redmond, WA: Tempus Book of Microsoft Press. Coffey, W., and R. Shearmur. (1998). "Factors and Correlates of Employment Growth in the Canadian Urban System, 1971- 1991." Growth and Change 29: 44 - 66. Crevoisier, O. (1993). Industrie et region: Les milieux novateurs de l'Arc jurassien. Neuchlitel, SW: EDS.


113

DeBresson, C. (1990). "L'analyse inter-industrielle et Ie changement technologique." Revue d'economie politique 100(6): 833 - 869. DeBresson, C., ed. (1996). Economic Interdependence and Innovative Activity: An InputOutput Analysis. Brookfield, VT: Elgar. Desrochers, P. (1998). "On the Abuse of Patents as Economic Indicators." Quarterly Journal ofAustrian Economics 1(4): 51-74. Desrochers, P. (1999). "De I'influence d'une ville diversifiee sur la combinaison de techniques: Typologie et analyse de processus." Ph.D. dissertation, Universite de Montreal, QC. Desrochers, P. (2001). "Geographical Proximity and the Transmission of Tacit Knowledge." Review ofAustrian Economics 14(1): 25 - 46. Economic Classification Policy Committee. (1993a). Aggregation Structures and Hierarchies. Issue Paper No.2. Washington, DC: Bureau of Economic Analysis, U.S. Department of Commerce. Economic Classification Policy Committee. (1993b). Conceptual Issues. Issue Paper No. l. Washington, DC: Bureau of Economic Analysis, U.S. Department of Commerce. Feldman, M. P. (1994). The Geography ofInnovation. Boston: Kluwer. Feldman, M. P. (2000). "Location and Innovation: The New Economic Geography of Innovation, Spillovers and Agglomeration." In G. L. Clark, M. P. Feldman, and M. S. Gertler, eds., Handbook of Economic Geography (pp.373 - 394). Oxford: Oxford University Press. Feldman, M. P., and D. B. Audretsch. (1999). "Innovation in Cities: Science-Based Diversity, Specialization and Localized Competition." European Economic Review 43: 402 - 429. Fores, M. (1979). "The History of Technology: An Alternative View." Technology and Culture 20(4): 853 - 860. Galunic, C., and S. Rodan. (1997). "Resource Recombination in the Firm: Knowledge Structures and the Potential for Schumpeterian Innovation." INSEAD Working Paper #97/89/0B/SM. INSEAD, Paris. Glaeser, E., H. Kallal, J. Scheinkman, and A. Shleifer. (1992). "Growth in Cities." Journal of Political Economy 100(6): 1126 - 1152. Glasmeier, A. (2000). "Economic Geography in Practice: Local Economic Development Policy." In G. L. Clark, M. P. Feldman, and M. S. Gertler, eds., The Oxford Handbook of Economic Geography (pp. 559 - 579). Oxford: Oxford University Press. pp. 559-579. Griliches, Z. (1990). "Patent Statistics as Economic Indicator: A Survey." Journal of Economic Literature 28(12): 1661 - 1707.

114


Harrison, B., M. ~elley, and J. Gant. (1996a). "Innovative Firm Behavior and Local Milieu: Exploring the Intersection of Agglomeration, Firm Effects, and Technological Change." Economic Geography 72(3): 233 - 258. Harrison, B., M. Kelley, and J. Gant. (1996b). "Specialization Versus Diversity in Local Economies: The Implication for Innovative Private Sector Behavior." Cityscape: A Journal ofPolicy Development and Research 2(2): 61 - 93. Henderson, J. V. (1997). "Externalities and Industrial Development." Journal of Urban Economics 42: 449 - 470. Henderson, J. V., A. Kuncoro, and M. Turner. (1995). "Industrial Development in Cities." Journal ofPolitical Economy 103(5): 1067 - 1090. Hoke, D. R. (1990). Ingenious Yankees: The Rise ofthe American System ofManufactures in the Private Sector. New York: Columbia University Press. Hounshell, D. (1991). From the American System to Mass Production. 1800 - 1932. 1984. Baltimore: Johns Hopkins University Press. Jackson, R. W. (1984). "An Evaluation of Alternative Measures of Regional Industrial Diversification." Regional Studies 18(2): 103 - 112. Jacobs, J. (1969). The Economy ofCities. New York: Random House. Jacobs, J. (1984). Cities and the Wealth ofNations. New York: Random House. Jewkes, J., D. Sawers, and R. Stillerman. (1969). The Sources of Invention. 1958. London: MacMillan. Klemm, F. (1959). A History of Western Technology. New York: Scribner. Kodama, F. (1992). "Technology Fusion and the New R&D." Harvard Business Review 70(3): 70 - 78. Koestler, A. (1969). The Act ofCreation 1923. 1964. London: Hutchinson. Kroeber, A. L. (1948). Anthropology. 1923. New York: Harcourt Brace Jovanovich. Langrish, J., M. Gibbons, W. G. Evans, and F. R. Jevons. (1972). Wealth from Knowledge: Studies ofInnovation in Industry. London: MacMillan. Levinthal, D. A. (1998). "The Slow Pace of Rapid Technological Change: Gradualism and Punctuation in Technological Change." Industrial and Corporate Change 7(2): 217 - 247. Lichtenberg, R. M. (1960). One-Tenth ofa Nation: National Forces in the Economic Growth ofthe New York Region. Cambridge, MA: Harvard University Press. Malmberg, A. (1996). "Industrial Geography: Agglomeration and Local Milieu." Progress in Human Geography 20(3): 392 - 403.

Urban Diversity and Intersectora/ Diffusion

115

Malmberg, A. (1997). "Industrial Geography: Location and Learning." Progress in Human Geography 21(4): 573 - 582. Miller, R., and M. Cote. (1987). Growing the Next Silicon Valley: A Guide for Successful Regional Planning. Lexington, MA: Lexington Books. Mokyr, J. (1990). The Lever of Riches: Technological Creativity and Economic Progress. New York: Oxford University Press. Porter, M. E. (2000). "Location, Clusters, and Company Strategies." In G. L. Clark, M. P. Feldman, and M. S. Gertler, eds., Oxford Handbook of Economic Geography (pp. 253 274). Oxford: Oxford University Press. pp. 253-274. Quigley, 1. M. (1998). "Urban Diversity and Economic Growth." Journal of Economic Perspectives 12(2): 127 - 138. Rosenberg, N. (1976). Perspectives on Technology. New York: Cambridge University Press. Rosenberg, N. (1994). Exploring the Black Box: Technology. Economics. and History. New York: Cambridge University Press. Rosenberg, N. (1996). "Uncertainty and Technological Change." In 1. C. Fuhrer and Jane Sneddon Little, eds., Proceedings of the "Technology and Growth" Conference. Boston: Federal Reserve Bank of Boston, Conference Series No. 40. Rothschild, M. (1990). Bionomics: Economy as Ecosystem. New York: Holt. Sabal, D. (1981). Patterns of Technological Innovation. Reading, MA: Addison-Wesley. Company. Scherer, F. M. (1982). "Inter-Industry Technology Flows in the United States." Research Policy 11: 227 - 245. Schmookler, J. (1966). Invention and Economic Growth. Cambridge, MA: Harvard University Press. Smiles, S. (1967). Industrial Biography: Iron Workers and Tool Makers. 1863. London: Kelley. Smith, C. S. (1982). A Search for Structure: Selected Essays on Art, Science and History. Cambridge, MA: MIT Press. Thomson, R. (1991). "Crossover Inventors and Technological Linkages: American Shoemaking and the Broader Economy, 1848 - 1901." Technology and Culture 32(4): 1018 -1046. Usher, A. P. (1966). A History of Mechanical Invention. 1929. Cambridge, MA: Harvard University Press.

116


Von Hippel, E. (1988). The Sources a/Innovation. New York: Oxford University Press. Weber, R. J. (1992). Forks, Phonographs, and Hot Air Balloons: A Field Guide to Inventive Thinking. New York: Oxford University Press. I

2

In a recent paper, Levinthal (1998: 218) defines speciation as the "separation of reproductive activity" but then argues that the "application of existing technological knowhow to a new domain of application" is a speciation event. The main characteristic of speciation, however, is that the new organisms created through such a process can no longer mate together to produce offsprings. In short: (I) not an innnovations are patentable; (2) not an patentable innovations are patented; (3) there are strong biases in the propensity to patent depending on the industry of origin, the size of the firm, and the type of invention; (4) there are important reliability problems in patent data; (5) some new patents prove to have an economic value, but most do not; (6) many patents are of a purely defensive nature; 7) patent requirements have evolved drasticany over time and geographical space (Desrochers 1998; Griliches 1990).